Multi-frame fractal

Why fractals

Markets are self-similar: strip the axis labels off a five-minute chart and a weekly one and you will struggle to tell them apart. The same rhythms of accumulation, expansion and exhaustion repeat at every scale. That is not an aesthetic coincidence — it is the signature of a system running the same mechanics at many sizes at once.

The field that studies this is complexity science. It does not treat a market as a series of numbers to be smoothed, but as a dynamical system of interacting participants — with memory, with feedback, and with stretches where it organises itself before coming apart again. The same mathematics physics uses to describe phase transitions and critical phenomena describes markets too: when they are stable, and when they have moved somewhere fragile.

That leads to a practical question ordinary indicators cannot answer: what kind of market is this right now? A breakout in a persistent regime is a continuation; the identical breakout in a mean-reverting one is something to fade. Same pattern on the chart, opposite conclusions — and what decides it is the regime, not the candle.

What BunnyQuant does differently

The classical fractal measures date from the middle of the last century and were built for very long, reasonably well-behaved series. Applied straight to financial data they run into two problems: markets do not hold still for long enough, and over a short window you can always compute a number, but not one worth trusting.

The toolset here was built by the author over years of research in complexity science and fractal analysis, then carried into finance — resting on dynamical systems and on methods that come out of statistical physics. Three things set it apart:

• Measured on two time scales. One measurement over a long window for the background picture, and an independent check over a short window to catch the present state. When the two disagree, that disagreement is itself information.
• It comes with a confidence, and it will refuse. Every figure carries an interval. When the data is too short or the measurement too unstable to trust, the system reports that it cannot tell rather than returning a confident-looking number. A wrong regime call is worse than none.
• Agreement across frames, as one number. It does not just score each timeframe and leave you to assemble them. It measures whether the frames are in phase or fighting each other, because structure that holds across scales is worth far more than a clean signal on a single frame.

This describes market structure; it is not a buy-and-sell signal generator. It tells you what kind of market you are standing in — the decision to trade remains yours.